Intracellular cytokines in blood T cells in lung transplant patients – a more relevant indicator of immunosuppression than drug levels

Abstract

Allograft rejection remains a major cause of morbidity and mortality following lung transplantation and is associated with an increase in T-cell pro-inflammatory cytokine expression. Systemic levels of immunosuppressive drugs used to reduce pro-inflammatory cytokine expression are closely monitored to their ‘therapeutic range’. However, it is currently unknown if levels of these drugs correlate with pro-inflammatory cytokine expression in peripheral blood T cells. To investigate the immunomodulatory effects of currently used immunosuppressive regimes on peripheral blood T-cell cytokine production, whole blood from stable lung transplant patients and control volunteers  were  stimulated in vitro and cytokine production by CD8+ and CD4+ T-cell subsets determined using multiparameter flow cytometry. T-cell IL-2 and TNFα production was significantly reduced from lung transplant patients compared to controls. CD4+ T-cell production of IFNγ was also significantly reduced from lung transplant patients but production of IFNγ by CD8+ T cells remained unchanged. There was an excellent correlation between the percentage of CD8+ T cells and the percentage of CD8+ T cells producing IFNγ from transplant patients. T-cell IL-4 and CD8+ T-cell production of TGFβ was significantly increased from lung transplant patients. We now provide evidence that current immunosuppression protocols have limited effect on peripheral blood IFNγ production by CD8+ T-cells but do up-regulate T-cell anti-inflammatory cytokines. Drugs that effectively reduce IFNγ production by CD8+ T cells may improve current protocols for reducing graft rejection in these patients. Intracellular cytokine analysis using flow cytometry may be a more appropriate indicator of immunosuppression than drug levels in these patients. This technique may prove useful in optimizing therapy for individual patients.

Introduction

Survival after lung transplantation is less than 50% after 5 years. Allograft rejection remains a major cause of morbidity and mortality following lung transplantation and is associated with an increase in pro-inflammatory cytokine expression in graft-infiltrating T-cells [1]. Acute rejection has been shown to be associated with an increase in IL-2 and other pro-inflammatory cytokines such as IFN-γ and TNF-α, while chronic rejection or obliterative bronchialitis (OB) is associated with a modest increase in pro-inflammatory cytokine production and also TGF-β[2,3].

The most effective transplantation immunosuppressive strategies are based on interruption of IL-2 signalling by calcineurin inhibitors, Cyclosporin A (CsA) and Tacrolimus (Tac). However, treatment with these drugs is associated with serious adverse side-effects including nephrotoxic, diabetogenic, neurological and cardiovascular effects [4]. Pharmacokinetic properties of both drugs show high inter- and intra-individual variability and both drugs have a narrow therapeutic index, necessitating therapeutic whole-blood monitoring to optimize treatment, i.e. therapeutic effectiveness is currently indirectly assessed by measurement of Tac and CsA levels in plasma. However, it is currently unknown if levels of these drugs correlate with pro-inflammatory cytokine expression in peripheral blood T cells.

Lymphocytes are known to traffic from the blood stream to the lung and later rejoin the peripheral circulation [5] suggesting that measurement of blood T cell cytokines may be reflective of graft infiltrating T-cell cytokine profiles.

Previous methodology to measure cytokine levels in transplant patients include ELISA quantification from serum and PBMC culture [6,7] and RTPCR of cytokine mRNA levels [1]. ELISA gives no indication of cytokine-producing cell types and is time consuming and expensive if several cytokines are quantified. PBMC purification techniques lead to increased apoptosis of cells as we have previously shown [8].

While RTPCR is a sensitive technique, results depend on purification of cells from heterogeneous cell populations and are subject to technical error.

We have previously measured intracellular pro- and anti-inflammatory T-cell cytokines using flow cytometry [9,10]. To investigate the immunomodulatory effects of currently used immunosuppressive regimes, whole blood from lung transplant patients and control volunteers were stimulated in vitro and T cell cytokine production determined using multiparameter flow cytometry.

Materials and methods

Patient and control groups

Nine lung transplant recipients with no clinical or histopathological evidence of current acute or chronic rejection or infection were invited to participate in the study and fully informed consent obtained (Table 1) following institutional ethics approval. All patients were tested to exclude infection including CMV (histopathologically, rapid viral culture and CMV PCR of BAL) and mycoplasma (enzyme immunoassay of BAL). Five of these patients were re-tested at their next routine surveillance bronchoscopy. All transplant patients were at least 3 months post-transplant. Details of patients and previous acute rejection episodes are presented in Table 2. Immunosuppression therapy comprised combinations of either CsA or Tac with prednisolone and azathioprine. Trough drug levels of either CsA or Tac were within recommended therapeutic ranges (Table 2). Fifteen healthy age-matched volunteers were recruited as controls. Venous blood was collected and added to 10 U/ml preservative free sodium heparin (DBL, Sydney, Australia) from the lung transplant patients and control volunteers and samples were processed within 2 h of collection.

Table 1.

Demographic details of the populations studied: Lung transplant patients and control subjects (mean ± SD).

	Control	Transplant
No. of subjects	15	9*
Age (years)	44 (± 8)	41 (± 13)
FEV1 (% pred)	100·4 (± 24)	72·3 (± 24)
FVC (% pred)	99 (± 14)	78·0 (± 14)
FEV1 (% FVC)	96·4 (± 26)	77·3 (± 17)

^*Nine patients, 15 episodes.

Table 2.

Demographic details of the populations studied: Lung transplant patients and previous acute rejection (ACR) episodes.

Patient	Predisposing pathology	Time post transplant (months)	Grade ACR	Number of prior ACR episodes	Grade of prior ACR episodes	CsA/Tac levels*
1	Cystic Fibrosis	6	A0B0	1	A2	Tac 14·5
1		12	A0B0	1	A2	Tac 11·4
2	Bronchiectasis	6	A0B0	0		Tac 9
3	Pulmonary hypertension	3	A0BX	0		CsA 276
3		9	A0B0	0		CsA 258
4	Congenital bronchial webbs	12	A0BX	2	A3B0, A2B0	CsA 300
						CsA 265
4		18	A0B0	2	A3B0, A2B0	CsA 260
						CsA 245
5	Emphysema	3	A0B0	0		Tac 6
6	Pulmonary hypertension	6	A0BX	0		CsA 349
7	Emphysema	6	A0B1	1	A2B0	CsA 152
7		9	A0B0	1	A2B0	CsA 205
8	Pulmonary hypertension	9	A0BX	0		CsA 235
8		9	A0B0	0		CsA 185
9	Bronchiectasis	9	A0B0	0		Tac 12

^*Therapeutic range for CsA (80–250 µg/l) and Tac (5–20 µg/l).

Leucocyte counts

Full blood counts, including white cell differential counts, were determined using a CELL-DYN 4000 (Abbot Diagnostics, Sydney, Australia). Blood films were stained by the May – Grunwald-Giemsa method and white cell differential counts checked by morphological assessment microscopically.

Absolute lymphocyte and CD4+ and CD8+ T-cell counts

One hundred microlitres of peripheral blood was stained with appropriately diluted fluorescently conjugated monoclonal antibodies to CD8 FITC (BD Biosciences, Sydney, Australia), CD4 PE (BD) and CD3 PC5 (Coulter/Immunotech, Florida, USA) as previously described [10]. Samples were analysed by gating using forward scatter (FSC) versus side scatter (SSC) to exclude platelets and debris. Gated cells were analysed with CD45/CD14 to ascertain that cells were of lymphoid origin as previously reported [9]. Control staining of leucocytes with anti-mouse IgG1-FITC/IgG1a-PE/IgG1-PC5 was performed on each sample and background readings of <2% were obtained. A minimum of 10 000 CD3 + cells was acquired in list mode format for analysis.

Leucocyte stimulation

One ml aliquots of blood were diluted 1 : 2 with RPMI 1640 medium (Gibco, NY, USA) supplemented with 125 U/ml penicillin and 125 U/ml streptomycin (Gibco) in 10 ml sterile conical PVC tubes (Johns Professional Products, Sydney, Australia). Phorbol myristate (25 ng/ml) (Sigma, Sydney, Australia) and ionomycin (1 µg/ml) (Sigma) was added for T-cell cytokine stimulation. Brefeldin A (10 µg/ml) was added as a ‘Golgi block’ (Sigma) and the tubes reincubated in a humidified 5% CO₂/95% air atmosphere at 37°C.

Cytokine determination

Cytokine determination was performed as previously reported [9]. Briefly, at 24 h 100 µl 20 m m EDTA/PBS was added to one of the whole blood culture tubes which was vortexed vigorously for 20 s to remove adherent cells. To lyse red blood cells, 2 ml of FACSlyse solution (BD) was added and tubes incubated for 10 min at room temperature in the dark. After centrifugation at 500× g for 5 min and decanting, 0·5 ml 1 : 10 diluted FACSperm (BD) was added to each tube, mixed, and incubated a further 10 min at room temperature in the dark. Two ml 0·5% bovine serum albumin (Sigma)/Isoton II (Beckman Coulter) was then added and the tubes centrifuged at 300× g for 5 min. After decanting supernatant, Fc receptors were blocked with 10 µl human immunoglobulin (Intragam, CSL, Parkville, Australia) for 10 min at room temperature. Five µl of appropriately diluted anti-CD8 (BD) and anti-CD3 PC5 (Coulter/Immunotech) PE-conjugated anti-cytokine monoclonal antibodies to IL2, IL4, IFNγ, TNFα (BD) and TGFβ (IQ Products, Groningen, Nederlands) or isotype control monoclonal antibody was added for 15 min in the dark at room temperature. Two mls of 0·5% bovine serum albumin (Sigma)/Isoton II (Beckman Coulter) was then added and the tubes centrifuged at 300× g for 5 min. After decanting, cells were analysed within 1 h on a FACSCalibur flow cytometer using CellQuest software (BD). Samples were analysed by live gating using FL3 staining versus side scatter (SSC). A minimum of 10 000 CD3 positive, low SSC events were acquired in list-mode format for analysis. Control staining of cells with anti-mouse IgG1-PE/IgG-PC5 was performed on each sample and background readings of <2% were obtained.

Statistical analysis

Statistical analysis was performed using the nonparametric Mann–Whitney and Pearson correlation tests using SPSS software and differences between groups of P < 0·05 considered significant.

Results

Absolute lymphocyte and CD4+ and CD8+ T-cell counts

There was no significant difference between the absolute leucocyte count for patient and control groups (7·9 ± 4·3 and 6·5 ± 2·3 × 10⁹/l, mean ± SD for patient and control group, respectively, P > 0·05). There was no significant difference in the absolute lymphocyte counts for patient and control groups (1·3 ± 0·6 and 1·3 ± 0·7 × 10⁹/l, mean ± SD for patient and control group, respectively, P > 0·05).

There was no significant difference in the absolute T lymphocyte count for patient and control groups (1·4 ± 1·2 and 1·5 ± 1·3 × 10⁹/l, mean ± SD for patient and control group, respectively, P > 0·05).

The percentage of CD4+ T cells was significantly reduced in the overall transplant compared to control group (53 ± 11·7 and 62 ± 6·9, mean ± SD for patient and control group, respectively, P = 0·048). The percentage of CD8+ T cells was significantly increased in the overall transplant compared to control group (47 ± 11·7 and 37 ± 8·0, mean ± SD for patient and control group, respectively, P = 0·048). The percentage of CD4-CD8- and CD4+ CD8+ T cells was <3% for all patient and control subjects.

One patient (Patient no. 1, a CMV negative recipient, CMV positive donor) subsequently developed CMV. Statistical significance for all data was unchanged when this patient was excluded from the analysis.

Intracellular pro-inflammatory T-cell cytokine production

There were several differences noted for intracellular T-cell cytokine production between lung transplant patients and control group.

The percentage of CD8- (CD4+) and CD8+ T-cells producing IL-2 and TNFα was significantly reduced from lung transplant patients compared to controls (Fig. 1). The reduction in the percentage of T cells expressing IL-2 and TNFα was more marked for CD4+ T cells than for CD8 + cells. The percentage of CD4+ T-cells producing IFNγ was also significantly reduced from lung transplant patients but percentage of CD8+ T cells producing IFNγ by remained unchanged (Fig. 1). A representative dot plot showing IFNγ production from CD8+ and CD8- (CD4+) T cells from two patients and control are shown in Fig. 2. There was an excellent correlation between the percentage of CD8+ T cells and the percentage of CD8+ T cells producing IFNγ (R = 0·995, P = 0·001). There was no other correlation between any other T-cell subset and cytokine in either the patient or control group. There was no correlation between decreased pro-inflammatory cytokine expression and CsA or Tac levels (data not shown).

Fig. 1.

Box plot graphs showing the production of pro-inflammatory cytokines by CD4+ (□) and CD8+ () T-cells from lung transplant (T) and control (C) subjects following in vitro stimulation (mean ± SD and range). The percentage of CD4+ and CD8+ T cells producing IL-2 and TNF-α was significantly reduced from lung transplant patients compared to control. The percentage of CD4+ T cells producing IFN-γ was also significantly reduced from lung tranplant patients but not the percentage of CD8+ T cells producing IFN-γ. Note the marked inhibition of inflammaatory T-cell cytokines in CD4+ cells compared with CD8+ cells in transplant patients compared to control group.

Fig. 2.

Representative dot plots showing the effective immunosuppression therapy on IFN-γ production by CD4+ and CD8+ T cells from 2 lung transplant patients and control. T cells were identified by CD3 PC5 versus side scatter characteristics. Patient A shows immunosuppression of IFN-γ in CD8–(CD4+) T cells but not in CD8+ T cells. Patient B shows immunosuppression of IFN-γ in both T cells subsets. Note the reduced percentage of CD4 T cells and increased CD8 T cells in transplant patients compared to control.

Intracellular anti-inflammatory T-cell cytokine production

The percentage of CD4+ T-cells producing IL-4 was significantly increased (P = 0·019) from lung transplant patients (1·0 ± 0·8% and 0·4 ± 0·3%, mean ± SD for transplant and control groups, respectively). The percentage of CD8+ T-cells producing IL-4 was significantly increased (P = 0·010) from lung transplant patients (1·4 ± 1·3% and 0·0 ± 0·0%, mean ± SD for transplant and control groups, respectively).

The percentage of CD8+ T-cells producing TGFβ was significantly increased (P = 0·017) from lung transplant patients (4·6 ± 2·4% and 3·0 ± 2·6%, mean ± SD for transplant and control groups, respectively). The percentage of CD4+ T-cells producing TGFβ was unchanged (P = 0·890) from lung transplant patients (4·9 ± 3·1% and 5·3 ± 3·2%, mean ± SD for transplant and control groups, respectively). There was no correlation between increased anti-inflammatory cytokine expression and CsA or Tac levels (data not shown).

Discussion

This is the first report of intracellular pro- and anti-inflammatory cytokines in peripheral blood T cells from lung transplant patients and provides important new information regarding the immunosuppressive effect of current drug protocols in these patients. Lymphocytes are known to traffic from the blood stream to the lung and later rejoin the peripheral circulation [5] suggesting that measurement of blood T-cell cytokines may be reflective of graft infiltrating T-cell cytokine profiles. We provide evidence that current immunosuppression protocols have limited effect on peripheral blood CD8+ T-cells, particularly IFNγ inflammatory cytokine production in lung transplant patients. All transplant patients in this study had plasma levels of CsA or Tac within their therapeutic ranges. Our findings therefore suggest that analysis of cytokine production may provide a more accurate assessment of immunosuppression than drug levels. Our results for T-cell IL-2 and IFNγ production are consistent with those of a previous study [11] from a variety of transplant patients. However, this previous study failed to measure T-cell TNFα production, an inflammatory cytokine reportedly elevated during transplant rejection [1], and also the important anti-inflammatory cytokines IL-4 and TGFβ.

In contrast to our results, a recent report failed to show any differences between inflammatory T-cell cytokines in stable renal transplant patients and controls [12]. However, this study did not distinguish between CD8+ and CD8- T-cell cytokine production and our present study shows that immunosuppression protocols are adequate for CD8- (CD4+) but not CD8+ T-cell inflammatory cytokines, particularly IFNγ.

Importantly, we have also shown that current immunosuppression agents increase anti-inflammatory cytokine production of IL-4 and TGFβ by T cells from lung transplant patients. IL-4 negatively regulates Th1 cytokines IFNγ and IL-2 [13] that have been reportedly increased during acute rejection episodes [1]. TGFβ has also been shown to inhibit IL-2 and IFNγ production by T cells [14]. Thus, our findings of increased IL-4 and TGFβ production by T-cells in lung transplant patients may partially explain the significantly reduced levels of IFNγ and IL-2 in peripheral blood T cells. The increased sensitivity of CD4+ T cells to TGFβ in reducing Th1 responses compared to CD8+ T cells [15] may help explain our observation of increased inhibition of these cytokines in CD4+ cells compared to CD8+ T cells.

Our previous report that IL-2 and IFNγ are inhibited in the presence of methylprednisolone [16,17] is consistent with the reported effects of CsA and Tac [18]. However, although Tac and  CsA  have  been  shown  to  inhibit  T-cell  IL-4  production in vitro[18], low levels of corticosteroids have previously been shown to be stimulatory for T-cell IL-4 production [19] and may be acting similarly in transplant patients.

The combined therapy of CsA or Tac and methylprednisolone may therefore account for the significant reduction in these pro-inflammatory T-cell cytokines in these patients.

Although total wcc and lymphocyte counts were unaltered, our finding of increased CD8+ T cells in transplant patients is consistent with a previous report [20]. TGFβ has been shown to be costimulatory for CD8+ T cells but not CD4+ T cells [21]. IL-4 enhances the proliferation of precursors of cytotoxic lymphocytes and their differentiation into active cytotoxic CD8+ T cells [22]. In CsA or Tac treated mice, T-cell proliferation was shown to be suppressed in CD4+ but not CD8+ subsets [23]. Our finding of increased TGFβ by CD8+ T cells and increased IL-4 production by both CD4+ and CD8+ T-cell subsets may therefore be causative factors in the significant increase in cytotoxic T cells in these patients. The relative increase in absolute numbers of CD8+ T cells and excellent correlation between the percentage of CD8+ T cells and the amount of IFNγ being produced by these cytotoxic cells suggests that current immunosuppressive protocols are ineffective at reducing this inflammatory cytokine. Chronic graft rejection is associated with increased fibrosis in the lung. Both IL-4 [24] and TGFβ[25] have been shown to promote fibroblast proliferation in the lung. Therefore, although TGFβ and IL4 are anti-inflammatory cytokines, their increased production by cytotoxic lymphocytes that migrate to the lung may contribute to increased fibrosis (as that observed in chronic graft rejection).

Current immunosuppression protocols are not without significant toxic side-effects [26]. In an attempt to minimize these side-effects a number of low toxicity protocols have been developed [27]. One could hypothesize that if a transplant patient was showing signs of drug toxicity and levels of intracellular T-cell inflammatory cytokines were markedly reduced compared to control (e.g., Patient B in Fig. 2), the dose of drug could be reduced and therapeutic effects monitored or tailored to suit the individual patient. Alternately, if intracellular T-cell inflammatory cytokines were not reduced (e.g., CD8 T cells in Patient A, Fig. 2), other immunosuppressive drugs should be considered.

In conclusion, we now provide evidence that current immunosuppression protocols have limited effect on peripheral blood IFNγ production by CD8+ T-cells but do up-regulate T-cell anti-inflammatory cytokines, TGFβ and IL4. Drugs that effectively reduce IFNγ production by CD8+ T cells may improve current protocols for reducing graft rejection in these patients.

Intracellular cytokine analysis using flow cytometry may be a more appropriate indicator of immunosuppression than drug levels in these patients. This technique may prove useful in optimizing therapy for individual patients.